Validation and simplification of Bethesda guidelines for identifying apparently sporadic forms of colorectal carcinoma with microsatellite instability

BACKGROUND Colorectal tumors with microsatellite instability (MSI) that do not comply with previously defined clinical criteria may be found in recently diagnosed hereditary nonpolyposis colorectal carcinoma families. Until recently, the indications for MSI testing were not clearly established. The objective of the current study was to validate the recently published Bethesda guidelines for MSI testing in a series of patients with apparently sporadic forms of colorectal carcinoma (CRC). METHODS Sixty‐two patients with so‐called sporadic CRC were included in the current study. MSI was analyzed at seven poly(CA) repeat sequences and at one poly(A) locus. RESULTS Nine of 62 patients (14.5%) had tumors exhibiting MSI at ≥ 2 loci and 7 patients (11%) had MSI at ≥ 3 loci. Patients with MSI positive tumors were younger (P < 0.05), and their tumors more frequently were right‐sided (P < 0.02) and more often exhibited a mucinous component (P < 0.05). The Bethesda guidelines were positive in 18% (11 of 62) of patients. The sensitivity of these guidelines in identifying tumors with MSI at ≥ 3 loci was 43% and the positive predictive value (PPV) was 27% (3 of 11 cases). Other variables were considered as alternative criteria to identify CRCs with MSI: age < 45 years and/or a right‐sided tumor with a mucinous component. Using these 2 criteria alone, sensitivity increased to 85% and PPV to 46%. CONCLUSIONS In this study group, the use of three clinical criteria as sole indicators for MSI testing in patients with apparently sporadic forms of CRC were significantly more discriminating compared with the Bethesda guidelines, in addition to being substantially easier. Cancer 1999;85:779–85. © 1999 American Cancer Society. In the current study the authors applied the Bethesda guidelines, which were developed to identify colorectal tumors with microsatellite instability (MSI), to a population of 62 patients with apparently sporadic forms of colorectal carcinoma. Although the guidelines were highly specific, their sensitivity and positive predictive value in the identification of tumors with MSI at ≥3 loci were only 43% and 27%, respectively. The authors propose a simplification of the Bethesda guidelines that increases the sensitivity to 85% and the positive predictive value to 46%.